Abstract: Students evaluation by traditional quiz is based on the answers, which is hard to show students answering process and cognitive strategies. Therefore, visual analytics system for eye tracking data in quiz, GazeQuiz, is proposed. The system collected students gazed at a during the process of answering quiz and analyzed the data by visualization. GazeQuiz showed students’ individual cognitive behavior through visualization graph such as transition diagram and scarf plot, and revealed group attention distribution through box plot and scatter plot. The teaching application proves that GazeQuiz can quickly and accurately analyze the student testing process, and can help teachers understand the individual and group characteristics of students’ testing process, which solves accurate positioning of learning problems.